I Field of Invention
The present invention relates to radio engineering, particularly, to methods and devices of multipath signal receiving in CDMA systems and can be used in BTS receiving equipment.
II Description of the Related Art
Today cellular communication systems are being developed at an amazing speed. Reduction of service cost and fast growth of the number of users are forced by the increasing demand for system capacity, capacity defined as a number of simultaneously served users per cell. In addition, new data exchange networks like Internet impose new requirements to data transmission rate and propagation channel reliability.
These requirements have accelerated development of signal processing methods and led to the emergence of new radio communication systems. Among latest achievements in this field is CDMA systems. There are CDMA based cellular systems currently operating according to IS-95 Mobile Station—Base Station Compatibility Standard for Dual—Mode Wideband Spread Spectrum Cellular System (to be published as IS-95).—Qualcomm Inc., 3 Volumes, March 1993.—2123 p. and there are third generation standards under development for future wireless networks: UMTS [The ETSI UMTS Terrestrial Radio Access (UTRA) ITU-R RTT Candidate Submission. (UMTS Standard)] and cdma2000 [The ETSI UMTS Terrestrial Radio Access (UTRA) ITU-R RTT Candidate Submission. (UMTS Standard)]. They are supposed to add new service functions, such as high rate channels, access to Internet, location, etc.
CDMA systems are asynchronous address systems, where signals from different users share a common frequency bandwidth and users are separated based on the signal type—a unique function, scrambling function, is assigned to each user. Since signals of different users arrive at the receiving end with random delays, it is not possible to provide full mutual orthogonality of signals from different users. Therefore, it is very critical to jointly differentiate and estimate signal parameters of all the users simultaneously processed at the receiving end. This task is referred to as “multi-user detection” [4] S. Verdu “Optimum Multiuser Asymptotic Efficiency”, IEEE Transactions on Communications, vol. COM-34, 9, Sep. 1986, pp. 890–897.
Despite the huge interest in multi-user detection: Z. XIE, R. T. Short, and G. K. Rushforth “A Family of Suboptimum Detectors for Coherent Multiuser Communication”, IEEE Journal on selected areas in communications, vol. 8, no. 4, May 1990, pp. 683–690, B. Wu, Wang, “New Sub-Optimal Multiuser Detectors for Synchronous CDMA Systems”, Proceedings Pacific Rim Conference on Communications, Victoria, BC, Canada, IEEE, May, 1995, Z. Zvonar, M. Stojanivic, “Performance of Multiuser Diversity Reception in Nonselective Rayleigh Fading CDMA Channels”, IEEE Personal Communications, 1994, pp. 171–175. etc. there are still a lot of unsolved issues. For example, there is a problem to develop simple and effective methods and devices of simultaneous receiving of signals of multiple users under the conditions of a priori unknown complex envelope of receiving multipath signals with several info channels per user.
At present there are different methods and devices of multipath signals in CDMA systems.
There is a method of signal receiving and CDMA communication system developed by Qualcomm according to the IS-95 standard “An Overview of Application of Code Division Multiple Access (CDMA) to Digital Cellular Systems and Personal Cellular Networks”, USA, Qualcomm, May, 1992, Document Number EX60-10010, where base (central) station, BS, comprises N receivers that receive signals from mobile stations. The level of structural interference at BS in this system is reduced because of the use of adaptive power control of mobile station signals.
However, the above method of signal receiving and communication system do not provide high interference immunity, power losses during signal receiving are possible because of the presence of multi-user interference.
There are methods and algorithms of multi-user detection in synchronous coherent system of CDMA communication system described by Peter Kempf in the paper “On Multi-User Detection Schemes for Synchronous Coherent CDMA Systems”, IEEE Vehicular Technology Conference, pp. 479–483, 1995.
In this paper several methods of multi-user detection are addressed. Let us consider one of them.
It is assumed that N users are served in a communication system. Data transmission rates of different users, the length of info symbols are the same. Each user has a single information, info, channel. The complex envelopes of signals from different users are assumed to be known and methods of their estimation are not considered. Signal propagation channel is one path. Analysis of the suggested algorithm is carried out without fading.
Unknown info parameters of users are estimated through L stages by sequential compensation of interfering effect of user signals on each other. At each of these stages the correlation responses of user signals, on which the final decision has been made at the previous stages, are formed. Out of them N/L maximum by module correlation responses are selected, on which the final decision about info parameters is made. The estimates of interfering effect of signals of these users are obtained and the output signal of this stage is generated by subtracting the obtained estimates from the output signal of the previous stage.
In the described method of multi-user detection forming of the estimate of interfering impact of user signals and subtraction of this estimate are carried out at high intermediate frequency, which makes it a difficult task to implement this method.
The use of the described method supposes the knowledge of complex envelopes of user signals and does not have the mechanism of their effective estimation. This renders it impossible to use this method in fading and invariable channel conditions.
The presence of only info signal per user does not correspond to the structure of user signals in today's radio systems, where several info channels and pilot channel are available.
Propagation channel is assumed to be one path
There is a method multi-user description in a CDMA communication system described by Andrew L. C. Hui and Khaled Ben Letaief “Successive Interference Cancellation for Multiuser Asynchronous DS/CDMA Detectors in Multipath Fading Links”, IEEE, vol. 46, 3, Mar., 1998, pp. 384–391.
In is assumed that N users are served in a communication system. Data transmission rates of different users, the lengths of info symbols, are the same. Each user has a single info channel. The complex envelopes of signals from different users are assumed to be known and their estimation methods are not considered. Propagation channel is multipath. Analysis of the algorithms is carried out under fading conditions.
It is assumed that the receiving equipment of base station recovers complex envelopes and user delays highly accurately; the method of how this is done is not specified.
The method is implemented in the following manner. The input signal is demodulated thus forming the correlation responses of all the paths of all the users at the output. The info parameters are estimated through sequential compensation of the interfering impact of user signals on one another through L stages. At each of L stages:                the correlation responses of all the paths of each user are combined forming soft decisions on info parameters of users;        the user with maximum by modulo soft decision and the final decision about his info parameter is made;        considering the matrices of cross-correlation the estimate of interfering effect of a given user on the correlation responses of user signals paths by which final decision has not been made at the previous stages is formed;        the correlation responses of this stage are formed by subtracting the obtained estimates of interfering effect from the correlation responses of the previous stage.        
The use of this method supposes the presence of accurate estimates of complex envelopes of user signals that cannot be obtained in practice because the processes of obtaining complex envelope estimates and information parameters are interrelated.
A single info channel per user does not meet the user signal structure in today's radio systems, where a number of info channels is assumed.
The algorithm supposes similar data transmission rates of all the users that does not correspond to real conditions.
At each stage the final decision is made by one user, so at the final stage the final decision by N−L users has to be made, which, in case N is much greater than L (N>>L), reduces interference stability of an estimate. When the number of stages L is a bit lower than the number of users N, the algorithm becomes more complex due to multiple stages.
Finally there is a multi-user detection method in the CDMA IS-95 system described by A. Duel-Hallen, J. Holtzman, Z. Zvonar in “Multiuser Detection for CDMA Systems”, IEEE Personal Communications, April 1995, pp. 46–57.
In this system N users are served. The length of info symbols of different users in this system is the same. A user is supposed to have a single info channel. The estimates of complex envelopes of signals from different users are derived by non-coherent estimation of info symbols of each user with subsequent accumulation of complex correlation responses of symbols correlated in accordance with the estimates obtained. User propagation channel is assumed to be multipath. Analysis of the considered algorithm is carried out in fading conditions.
The mentioned method is implemented in the following manner. The input signal is demodulated thus forming the correlation responses of signals of all the paths of all the users at the output. The info parameters are estimated by serial compensation of the interfering effect of users on each other through N stages. Within each of N stages:                the correlation responses of signals of all the paths of each user are combined thus forming soft decisions about info parameters of user signals;        
the user of max by modulo soft decision is chosen and final decision about its info parameter is made;
considering the cross-correlation matrices the estimate of interfering effect of a signal from a given user on the correlation responses of signals of user paths by which the final decision at the previous stages has not been made is formed;
the correlation responses of this stage are formed by subtracting the obtained estimates of interfering effect from the correlation responses of the previous stage.
The method of estimation of complex envelopes of user signals used in the described algorithm is, first, limited by the IS-95 standard frames, second, is not so efficient for it does not consider the mutual effect of user signals on each other.
A single info channel per user does not correspond to the 3G user signal structure (IS-2000, UMTS, 3GPP), where a number of info channels are supposed to be available.
The method supposes the same length of info symbols of different users in this system, which does not correspond to the requirements of mobile 3G standards.
With a great number of users N owing to multiple stages implementation of the method becomes a complex tasks.